From: Namhyung Kim <namhyung@kernel.org>
To: Milian Wolff <milian.wolff@kdab.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>,
Jin Yao <yao.jin@linux.intel.com>,
"linux-kernel@vger.kernel.org" <Linux-kernel@vger.kernel.org>,
linux-perf-users <linux-perf-users@vger.kernel.org>,
Arnaldo Carvalho de Melo <acme@redhat.com>,
David Ahern <dsahern@gmail.com>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
kernel-team@lge.com
Subject: Re: [PATCH v2 11/14] perf report: cache srclines for callchain nodes
Date: Thu, 10 Aug 2017 23:56:24 +0900 [thread overview]
Message-ID: <CAM9d7chNxx8_Z10dViFtDbykpH+zXdhu44AAvpxD2=6wLz_Vkg@mail.gmail.com> (raw)
In-Reply-To: <4512331.57hptoN48J@agathebauer>
On Thu, Aug 10, 2017 at 8:51 PM, Milian Wolff <milian.wolff@kdab.com> wrote:
> On Donnerstag, 10. August 2017 04:13:25 CEST Namhyung Kim wrote:
>> Hi Milian,
>>
>> On Sun, Aug 06, 2017 at 11:24:43PM +0200, Milian Wolff wrote:
>> > On one hand this ensures that the memory is properly freed when
>> > the DSO gets freed. On the other hand this significantly speeds up
>> > the processing of the callchain nodes when lots of srclines are
>> > requested. For one of my data files e.g.:
>> >
>> > Before:
>> > Performance counter stats for 'perf report -s srcline -g srcline --
> stdio':
>> > 52496.495043 task-clock (msec) # 0.999 CPUs utilized
>> >
>> > 634 context-switches # 0.012 K/sec
>> >
>> > 2 cpu-migrations # 0.000 K/sec
>> >
>> > 191,561 page-faults # 0.004 M/sec
>> >
>> > 165,074,498,235 cycles # 3.144 GHz
>> > 334,170,832,408 instructions # 2.02 insn per
>> > cycle
>> >
>> > 90,220,029,745 branches # 1718.591 M/sec
>> >
>> > 654,525,177 branch-misses # 0.73% of all
>> > branches
>> >
>> > 52.533273822 seconds time elapsedProcessed 236605 events and lost 40
>> > chunks!>
>> > After:
>> > Performance counter stats for 'perf report -s srcline -g srcline --
> stdio':
>> > 22606.323706 task-clock (msec) # 1.000 CPUs utilized
>> >
>> > 31 context-switches # 0.001 K/sec
>> >
>> > 0 cpu-migrations # 0.000 K/sec
>> >
>> > 185,471 page-faults # 0.008 M/sec
>> >
>> > 71,188,113,681 cycles # 3.149 GHz
>> >
>> > 133,204,943,083 instructions # 1.87 insn per
>> > cycle
>> >
>> > 34,886,384,979 branches # 1543.214 M/sec
>> >
>> > 278,214,495 branch-misses # 0.80% of all
>> > branches
>> >
>> > 22.609857253 seconds time elapsed
>> >
>> > Note that the difference is only this large when `--inline` is not
>> > passed. In such situations, we would use the inliner cache and
>> > thus do not run this code path that often.
>> >
>> > I think that this cache should actually be used in other places, too.
>> > When looking at the valgrind leak report for perf report, we see tons
>> > of srclines being leaked, most notably from calls to
>> > hist_entry__get_srcline. The problem is that get_srcline has many
>> > different formatting options (show_sym, show_addr, potentially even
>> > unwind_inlines when calling __get_srcline directly). As such, the
>> > srcline cannot easily be cached for all calls, or we'd have to add
>> > caches for all formatting combinations (6 so far). An alternative
>> > would be to remove the formatting options and handle that on a
>> > different level - i.e. print the sym/addr on demand wherever we
>> > actually output something. And the unwind_inlines could be moved into
>> > a separate function that does not return the srcline.
>>
>> Agreed. Also I guess no need to unwind anymore to get a srcfile for
>> an entry with your change.
>
> Does this mean I should respin the patch series with the above changes
> integrated? Or can we get this in first and then continue with the cleanup as
> described above later on?
Nop, it can be done later IMHO. I will try to review the code next week.
Thanks,
Namhyung
next prev parent reply other threads:[~2017-08-10 14:56 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-08-06 21:24 [PATCH v2 00/14] generate full callchain cursor entries for inlined frames Milian Wolff
2017-08-06 21:24 ` [PATCH v2 01/14] perf report: remove code to handle inline frames from browsers Milian Wolff
2017-08-07 15:07 ` Arnaldo Carvalho de Melo
2017-08-07 19:22 ` Milian Wolff
2017-08-07 20:16 ` Arnaldo Carvalho de Melo
2017-08-06 21:24 ` [PATCH v2 02/14] perf util: take elf_name as const string in dso__demangle_sym Milian Wolff
2017-08-07 15:10 ` Arnaldo Carvalho de Melo
2017-08-14 17:48 ` [tip:perf/core] perf util: Take " tip-bot for Milian Wolff
2017-08-06 21:24 ` [PATCH v2 03/14] perf report: create real callchain entries for inlined frames Milian Wolff
2017-08-16 7:53 ` Namhyung Kim
2017-08-20 20:57 ` Milian Wolff
2017-08-28 12:18 ` Namhyung Kim
2017-09-06 13:13 ` Milian Wolff
2017-09-07 14:58 ` Namhyung Kim
2017-09-07 15:05 ` Milian Wolff
2017-09-07 15:16 ` Namhyung Kim
2017-08-06 21:24 ` [PATCH v2 04/14] perf report: fall-back to function name comparison for -g srcline Milian Wolff
2017-08-06 21:24 ` [PATCH v2 05/14] perf report: mark inlined frames in output by " (inlined)" suffix Milian Wolff
2017-08-06 21:24 ` [PATCH v2 06/14] perf script: mark inlined frames and do not print DSO for them Milian Wolff
2017-08-06 21:24 ` [PATCH v2 07/14] perf report: compare symbol name for inlined frames when matching Milian Wolff
2017-08-06 21:24 ` [PATCH v2 08/14] perf report: compare symbol name for inlined frames when sorting Milian Wolff
2017-08-06 21:24 ` [PATCH v2 09/14] perf report: properly handle branch count in match_chain Milian Wolff
2017-08-06 21:24 ` [PATCH v2 10/14] perf report: cache failed lookups of inlined frames Milian Wolff
2017-08-06 21:24 ` [PATCH v2 11/14] perf report: cache srclines for callchain nodes Milian Wolff
2017-08-10 2:13 ` Namhyung Kim
2017-08-10 11:51 ` Milian Wolff
2017-08-10 14:56 ` Namhyung Kim [this message]
2017-08-10 17:58 ` Arnaldo Carvalho de Melo
2017-08-11 11:28 ` Milian Wolff
2017-08-06 21:24 ` [PATCH v2 12/14] perf report: use srcline from callchain for hist entries Milian Wolff
2017-08-06 21:24 ` [PATCH v2 13/14] perf util: do not consider empty files as valid srclines Milian Wolff
2017-08-07 15:21 ` Arnaldo Carvalho de Melo
2017-08-14 17:48 ` [tip:perf/core] perf srcline: Do " tip-bot for Milian Wolff
2017-08-06 21:24 ` [PATCH v2 14/14] perf util: enable handling of inlined frames by default Milian Wolff
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CAM9d7chNxx8_Z10dViFtDbykpH+zXdhu44AAvpxD2=6wLz_Vkg@mail.gmail.com' \
--to=namhyung@kernel.org \
--cc=Linux-kernel@vger.kernel.org \
--cc=a.p.zijlstra@chello.nl \
--cc=acme@kernel.org \
--cc=acme@redhat.com \
--cc=dsahern@gmail.com \
--cc=kernel-team@lge.com \
--cc=linux-perf-users@vger.kernel.org \
--cc=milian.wolff@kdab.com \
--cc=yao.jin@linux.intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).